TB: Expected Value

Expected value is the probability-weighted average of all possible outcomes. To compute it: multiply each possible outcome by its probability, then sum. A decision is good if its expected value exceeds the alternatives — regardless of which specific outcome occurs.

Expected value separates the quality of a decision from the quality of its outcome. A decision with high expected value can produce a bad outcome (bad luck), and a low-EV decision can produce a good outcome (good luck). Consistently making high-EV decisions is the only reliable path to good outcomes over time. The error of judging decisions by outcomes — "resulting" — is the most common mistake in hindsight evaluation.

• AI: A Modern Approach: rational agent decision-making is formalized as expected utility maximization — the rational agent selects the action with highest expected utility given its beliefs.
• How to Decide: Duke's framework formalizes expected value reasoning for everyday decisions — map outcomes, assign probabilities, weight by payoff.

• Free Roll
• Hard Decisions Are Easy
• Decisions Stacking